Is God a Non-Explanation?

One commonly hears the complaint that God as a concept is not explanatorily useful, because it is too all-encompassing, speculative, or it is compatible with any situation. This objection may also be framed that God is a bad scientific hypothesis because God does not make any clear empirical predictions, or something of this sort. Connected to this complaint is the idea that God is a postulate of speculative reason. Speculative reason is non-empirical reason – reason that does not have direct empirical corroboration (the converse of which is empirical reason or reason with direct empirical corroboration). A suspicion and disparagement of speculative reason, particularly as it relates to religion, is often found in the ideas of atheists. The implication is that natural theology is not valid because they are products of speculative reason, even though arguments against God are invalid then for the same reason (because responses to theistic arguments also rely on speculative reason). Similarly, Sean Carroll claims that theism is not “well-defined”, and explicitly justifies this by saying that theism and natural theology does not behave like scientific theories – it is not quantifiable and doesn’t make clear empirical predictions.[i]  Peter Atkins makes claims in a similar vein. “And the invocation of a God as an explanation of anything… is in fact the apotheosis of laziness. It’s well suited to armchair brains, who prefer to indulge in adipose arguments. Now the invocation of God as an explanation of anything, is an admission of defeat and of ignorance, disguised as a pretense of understanding.”[ii]  Bertrand Russel’s celestial teapot or popular atheists’ notion of the “flying spagetthi monster” implies similar complaints about God as an explanation.

If God as a hypothesis is not well-defined, then it becomes ad hoc; an omnipotent God is compatible with almost any state of affairs. You can explain anything with it, and if you run into trouble, you can just modify it appropriately. Because you only have to rely on speculative reasoning (and not empirical corroboration as well), you can respond to almost any potential objection ex post facto (making God as an explanation, unfalsifiable and ad hoc). So it is implied that naturalism as a metaphysical system relies on empirical reason, while natural theology relies only on speculative reason without empirical corroboration. So, we have a number of different, but interconnected, complaints here: God, as an explanation of anything, is ad hoc or not well-defined, it is a result of suspect speculative reason, and partly as a result of the last two, it is also unfalsifiable. All of this is really a veiled (but weaker) form of logical positivism, which claims that something cannot be meaningful without empirical corroboration. Reason cannot be valid unless it has direct empirical corroboration. If speculative reasoning is always suspect, then this implicitly affirms a verificationist criterion. The claim made by these atheist advocates is somewhat weaker, but just as irrational, because all the usual criticisms of logical positivism still apply to it. (Refer to the article “Is God-talk meaningless” for a more complete refutation of logical positivism.) The claim is that speculative reason cannot get us to truth. Only empirical reason (reason with direct empirical corroboration) can get us to truth. But the claim that speculative reasoning is not true because it doesn’t have empirical corroboration, is self-defeating, because it also has no empirical corroboration. In order to give empirical corroboration to this principle, you would need to assume that it is true first, in order to establish it. In other words, you would need to assume that only empirical reason can get you to truth in order to establish the truth of this principle, which is circular. Other objections against logical positivism work just as well against the claim that empirical reason is the only game in town, such as the existence of common sense beliefs that do not have empirical corroboration (the reality of the external world, the existence of other minds, the existence of objective morality, even the existence of the first-person experience of consciousness). None of these things have empirical corroboration, but we are reasonable to believe them.

Logical positivism claims that something is not meaningful unless we have empirical corroboration. This objection against natural theology claims that something cannot be considered true unless it has empirical corroboration. So this is also a form of positivism but a different and weaker form of it. Think for a moment of how often we hear the notion that there must be scientific evidence for God’s existence, before we can accept his existence. This is called scientism: you cannot know anything unless it has been proved by scientific means. What does that mean? It means that something cannot be considered true unless it is empirically verified. Strictly speaking, this is empiricism, not logical positivism ( which is one form of empiricism), but I regard them as pretty much the same, because all of the same objections raised against logical positivism can be raised, with equal force, against empiricism. For example, the popular British comedian, Ricky Gervais, proclaims in an article for the Wall Street Journal , “I don’t believe in God because there is absolutely no scientific evidence for his existence, and from what I’ve heard, the very definition is a logical impossibility in this known universe.”[iii]

The implication is that in order for something to be true, in order for it to be rational to believe it, you must have scientific evidence for it. This claim, construed as an argument against God’s existence, is both circular and self-defeating. It is circular because it presupposes Gervais’ own metaphysical worldview: naturalism and its closely associated epistemological claim of scientism. It is only the case that you have to give scientific evidence for something in order to know it, if scientism is true. So Gervais is presupposing his own metaphysics in order to argue for it (the Bible is true because the Bible says so!). There must be scientific evidence for God’s existence because science is metaphysics. Gervais’ statement is also self-defeating, because the claim that you need to have scientific evidence in order to know anything at all, does not and cannot have scientific evidence. As we’ve already covered, the scientific method cannot logically have scientific evidence, because that would require circular argumentation. In order to confirm science scientifically, you would need to assume the truth of the scientific method in order to prove the truth of the scientific method, which is clearly illogical. So, scientism is self-defeating. (In other words, the claim that God must have scientific evidence or that God should be considered as a scientific theory, in order to be considered true, is an inference from an incoherent epistemology). In the same way, the contention that a truth-claim can’t be considered true unless we have empirical evidence to support it, is also incoherent, for the same reason. That claim in itself cannot have empirical support. And we have more positive reasons to believe this is false. As I’ve already pointed out, if this principle were true, we would have to disbelieve objective morality, the reality of the external world, the existence of other minds, rational intuitions (including mathematical intuitions, and the scientific method itself). All the objections that apply to logical positivism, apply more broadly to empiricism as well.

In addition, the complaint that God is a non-explanation because it is too all encompassing or too broad can also be applied to naturalism. Atheists like to compare natural theology to science, but this is not the correct analogue. Science is not, by itself, a metaphysics like theism. Naturalism is a metaphysics and is therefore the correct analogue.  Now, we have to be clear about what we are talking about. We are not talking about individual naturalistic explanations – these are not inherently metaphysical since most of them can be accepted by both naturalists and theists. Naturalism, as a metaphysics, or metaphysical naturalism, is the claim that only nature exists. That is what we’re talking about. Naturalism, then, is just as all-encompassing as theism. It does not make clear, specific, empirical predictions and it is not quantifiable as a claim about the universe. It only makes the prediction that nothing will have a reasonable supernatural explanation. But theism, conceived of as a philosophical hypothesis, predicts that at least some things in reality will have a reasonable supernatural explanation. So what’s the difference in terms of theoretical integrity? But it gets even worse for naturalism.

More fundamentally, metaphysical naturalism is also a product of speculative reason, since it cannot be verified empirically. This is different from saying that naturalistic explanations can be verified empirically. Naturalism as a metaphysical claim is more ambitious than any single naturalistic explanation – it claims that only nature exists. This cannot be verified empirically. It may be contended that empirical evidence can be proffered for naturalism indirectly (something I’ll deny in a little bit), but empirical evidence can also be proffered for the existence of God indirectly ( fine-tuning, cosmic beginning and contingency, historical evidence, etc.). So God is no more a postulate of speculative reason than is naturalism. But even further, I’ll claim that there is not even indirect empirical evidence for naturalism. Naturalism is the contention that only the natural exists. So can there be empirical evidence that supports this contention? Can there be empirical evidence specifically for the claim that only nature exists. It seems clear to me that there cannot be ( or at least that no such evidence exists). All the empirical evidence that you could offer to support naturalism is only evidence for the existence of nature, not that only nature exists. So, you cannot show that nature is all there is by appealing to empirical evidence (at least not any empirical evidence that we have). This is another respect in which naturalism is self-defeating. If only nature exists, then all truth should be discoverable through the scientific method (scientism). The scientific method establishes something as true through empirical evidence. Metaphysical naturalism does not have any empirical evidence, which means that it is false according to scientism (and therefore according to metaphysical naturalism).  The only empirical evidence you can give for the claim that only nature exists is perhaps that there is no empirical evidence that there is anything beyond nature. This is false, but let’s grant it for the sake of argument. So, the fact that there is no evidence that contradicts naturalism implies that naturalism is true? How is this rational? How can atheists then give some Christians a hard time for believing that lack of evidence disproving God’s existence establishes God’s existence? These atheists are doing exactly the same thing! Naturalism cannot be established by the fact that there is no evidence contradicting naturalism. This is a logical fallacy called argumentum ad ignorantiam or argument from ignorance. You cannot establish that y doesn’t exist or that only x exists, based on the fact that we don’t have any evidence for y. The absence of empirical evidence for something beyond nature cannot be empirical evidence for nothing beyond nature. You might say that you can make this leap because you have independent reason to think that naturalism is true. But you don’t. As we’ve already covered, naturalism is a completely different claim from the claim that the natural world as such exists. You cannot use evidence in favour of the latter in favour of the former. You cannot use evidence in favour of the existence of the natural world to support the claim that only nature exists. So we’ve established that while God can claim indirect empirical evidence (by appeal to cosmic beginning, fine-tuning, cosmic contingency etc.) naturalism does not have any empirical evidence, not even indirect empirical evidence.

Is Natural theology ad hoc?

Sean Carroll, in his debate with William Lane Craig, lists a number of objections against the existence of God and then contends, “At the end of the day, in theism, you basically expect the universe to be perfect. Under naturalism, it should be kind of a mess. This is very strong empirical evidence. Now I know what you’re thinking. You’re thinking, ‘But I can explain all of that.’ I know you can all – so can I. It’s not hard to come up with ex post facto explanations for why God would have done it that way. Why is it not hard? Because theism is not well-defined…”[iv] So Carroll claims that any responses you could give to his arguments against theism are ex post facto justifications. But what is to stop me from claiming that naturalists are involved in ex post facto reasoning when they respond to theistic arguments against naturalism? For example, naturalists sometimes use some fairly speculative theories of mind  (such as functionalism) to respond to Plantinga’s evolutionary argument against naturalism. What about the difficulty naturalism has with objective morality? Naturalism also has a difficulty in accounting for the mind, since truly naturalistic theories of mind end up eliminating what is self-evidently a reality: the first-person experience of consciousness. Also, naturalists use implausible scenarios to explain compelling miracle claims. I can take all of these instances and say, like Carroll, that naturalists are involved in ex post facto or ad hoc maneuvering to try force reality to fit their views. Secondly, it is possible to respond to Carroll’s arguments by appeal to the Scripture. This means that it cannot be ex post facto or ad hoc maneuvering, because Scripture was not written in anticipation of modern atheism. The problem is that Carroll is applying scientific theoretical virtues to metaphysical systems. All metaphysical systems are guilty of ad hocness in a sense, because all of them (including naturalism) attempt to give a complete view of reality ( and are therefore not “well-defined” in a scientific sense). Atheist scientists like Carroll are still in the grip of logical positivism – science is metaphysics. We‘ve already looked at the many reasons why this is a mistaken view. Also, if those objections that Carroll raised were really formal contradictions like a square circle, then theists should not be able to come up with any coherent response, no matter how hard they think. You can think until you’re blue in the face, you’ll never be able think a square circle into logical coherence. Carroll may respond that this is because a square and a circle are well-defined, but we’ll get to that in a second. It is very easy to come up with scenarios which make God compatible with the factoids that Carroll mentioned, because they are not even close to showing a logical contradiction or even a more probabilistic contradiction between God’s attributes and what we would expect from the world.

It is not exactly clear what Carroll means by “well-defined.” (His concept of being well-defined is not well-defined). He seems to mean something like specificity, quantifiability, being able to make predictions, and so forth.  In other words, metaphysical theories must be identical to scientific theories (i.e. logical positivism). However, Carroll can’t have it both ways. He can’t claim that the empirical features he mentions disconfirms God, and then claim that God does not make any clear predictions. This is a contradiction. If God does not make any clear empirical predictions, then it shouldn’t be possible to disconfirm God’s existence based on the empirical features of the world. So, Carroll argues that theism is not well-defined, but then theism becomes well-defined when he criticizes it, and then it stops being well-defined again when theists attempt to respond to those criticisms. But is naturalism “well-defined” in this way? We’ve already seen that if naturalism is a scientific theory, it is a very poor one. It does not ( and probably cannot )have scientific and empirical evidence. It does not make any clear predictions, except perhaps for the negative claim that there aren’t any supernatural entities. It is also not quantifiable at all. This is why metaphysics cannot play by the rules of physics – dealing with first principles of reality is, theoretically, a different ball game from dealing with the rest of reality. But this notion of what constitutes a good scientific theory is not itself well-defined by Carroll’s standards either. It is itself not quantifiable and it does not make specific predictions, and it is not very specific either. “Simplicity” is a theoretical virtue of science, but one can debate to and fro about what type of or level of simplicity a scientific theory should have, or what level of ad hocness and predictiveness is necessary or permissible, without getting anywhere. There is no empirical confirmation for questions for this, because any opinion you would have about it cannot be confirmed empirically without presupposing it. You cannot show that scientific theories are mostly correct when they display this level of simplicity, when that level of simplicity is included in your definition of scientific correctness, which it needs to be in order to be significant.

But are the objections Carroll raises even good? Firstly, on naturalism, we would not expect there to be an orderly world at all. There are many more ways for there to be disorder than order. Thus, without a superintending intelligence, a naturalistic universe should probably not have laws (which may preclude any sort of universe from existing, much less a life-permitting one). Secondly, the world can be kind of a mess on theism depending on what type of theism you are thinking of. There is no reason why the world would not be kind of a mess on Deism, if God is indifferent about the world. Also, Christian theism is compatible with a messed up world as long as there are free agents who do stupid things in that world. This response as well as other responses to the problem of evil don’t actually modify the traditional concept of God. You can’t assume some doctrines of Christianity and reject others just so that your argument works. You can’t assume the attribute of God’s love and then reject the doctrine of original sin, and other attributes of God such as justice and retribution, because then you’re not really talking about a Christian theism. This is the very definition of ad hoc argumentation. When you assume all of the Christian metaphysics (and not just the aspects of it that make your argument work), there is no reason why the world should be perfect at all. Not even close. This is not ad hoc or ex post facto reasoning. I have not modified Christian theology in order to respond to Carroll and that theology is based in the Bible and has been believed for two thousand years and was not formed in anticipation of atheist critiques.

Is Christianity unfalsifiable?

So is God or Christianity unfalsifiable? If you could show that the Bible is incorrect with regard to central Christian doctrines, you will have disproved Christianity. If something is truly unfalsifiable, then it shouldn’t be possible to give evidence against it. Also, if you could show that the resurrection probably didn’t occur through historical investigation, you will have falsified or at least dealt a major blow to Christianity. (Of course, this involves more than saying that there is no evidence for the resurrection. You need positive reason to suspect that it didn’t occur if you want to falsify it). If God were unfalsifiable there shouldn’t exist any plausible atheist arguments, such as the problem of evil, divine hiddenness, inconsistent revelation and incoherence of theism. Obviously, if God were unfalsifiable and if every scenario were compatible with the existence of God, then no plausible atheist arguments should exist. There should be nothing that prima facie contradicts theism. Of course, atheist arguments do ultimately fail, but they have rational plausibility. No doubt atheists believe those arguments are successful. But you cannot have it both ways. Either the concept of God is unfalsifiable or arguments against him succeed. Both cannot be true, because if the arguments succeed then the concept of God is clearly falsifiable. (This is similar to Carroll’s error mentioned earlier). Now, in popular atheism, one hears the claim that one cannot prove a negative and therefore cannot prove that atheism is true or give evidence in favour of it. But one never hears this sort of claim from academic philosophers. I can give an inductive proof that there are no pencils in the drawer next to me. I can give an inductive proof that there are no carpets in a particular room. You might say that the carpets or the pencils might be invisible. Sure, they might be. But I said an inductive proof, not a deductive one. In order to have a successful inductive proof, you don’t need to rule out every single logically possible scenario incompatible with your claim, you merely need to show that your claim is most probable. So these are inductive proofs. I can give a deductive proof that no square circles exist, because the concept is formally contradictory. So, it is certainly not true that you cannot prove a negative.

Skepticism and Unfalsifiability

Since atheists appeal to the fact that believers involve themselves in ad hoc maneuvering in order to get out of objections, I’ve included this section to show that much the same complaint can be levelled at them. It is interesting that the skepticism of the supernatural displayed by atheists is also often effectively unfalsifiable. The only way to falsify naturalism empirically is to show that some empirical phenomenon is better explained by appeal to supernatural causes.Yet, this is made difficult by the way naturalists typically respond when they are confronted with an alleged miracle where naturalistic explanations are awkward and implausible. They respond that it is some anomaly, or it represents some part of nature that we don’t understand yet. Natural laws were not broken, it simply represents some natural law that we haven’t discovered yet. Or maybe they were hallucinating at exactly the right moment etc. In his address at the Oxford Union debate on the existence of God, Peter Millican said that even if Jesus rose from the dead, this doesn’t imply that God exists.[v]  “One point John Lennox made, appealing to the resurrection, was to say that if the resurrection is true, there’s a God. That doesn’t follow at all. All that follows at most, is that there is some supernatural being capable of performing a physical resurrection on a human being. Why should that be omnipotent? Why should that be omniscient? Why should that be perfectly good? It could be a being who’s doing it in order to dupe loads of us into having belief in a false religion.” Millican’s comments here really call to mind Luke 16:30: “‘If they do not listen to Moses and the prophets, they will not be persuaded even if someone rises from the dead.” In a debate with David Wood, Michael Shermer makes his skepticism about God perfectly unfalsifiable by saying that anything we can attribute to God is indistinguishable from a sufficiently advanced extra-terrestrial intelligence.[vi] Ironically, Shermer spends a good part of his first speech arguing that religion is unfalsifiable. The resurrection of Jesus might not be deductive proofs of God’s existence, but why do these atheists suddenly require that when dealing with the existence of God, and not with the truth of scientific theories? Jesus’s resurrection is strong inductive evidence for the existence of God. It might not show all of the attributes of God, but it seems most likely that God or a being very much like him, is what is responsible for the resurrection. If you are always going to explain away events that really do seem like miracles perpetrated by supernatural causes with contrived naturalistic explanations, then metaphysical naturalism is not empirically falsifiable. If, as Peter Millican says, the resurrection of Jesus does not even establish the existence of the Christian God, then no empirical proof can, because empirical evidence can only establish something inductively. It is impossible to give a deductive proof of the existence of something from historical evidence.

Naturalists also constantly invent speculative scenarios to explain features of the world that don’t gel with naturalism (scenarios just as speculative as theistic responses to atheist arguments). This includes how naturalists explain miracle claims with good evidence ( like the resurrection), as well as consciousness, the beginning of the universe, the existence of natural laws, moral value and obligation. All of these at least resist physical explanation. One can also not respond that a supernatural explanation will always be at least as improbable as an improbable naturalistic explanation. This is because, as we’ve already covered, there is no independent reason to believe that naturalism is true, while there are independent reasons to believe that God exists. Naturalism has no empirical evidence, not even indirect empirical evidence, while the God hypothesis does have indirect empirical evidence. Naturalism doesn’t have to be true for a naturalistic explanation to be more appropriate than a supernatural explanation. However, it does mean that an improbable or implausible naturalistic explanation will not automatically be preferable to a supernatural explanation.

[i] “William Lane Craig and Sean Carroll, ‘God and Cosmology’, 2014 Greer Heard Forum” Youtube Video, posted by ReasonableFaithOrg, March 3, 2014, 42:00

[ii] “What is the evidence for/against the existence of God?” Youtube Video, posted by drcraigvideos, December 15, 2010, 29:44

[iii] Ricky Gervais, “Ricky Gervais: Why I’m an Atheist”, Wall Street Journal, Dec 19, 2010, accessed March 23, 2017,

[iv] “William Lane Craig and Sean Carroll, ‘God and Cosmology’, 2014 Greer Heard Forum” Youtube Video, posted by ReasonableFaithOrg, March 3, 2014,     47:26

[v] “Professor Peter Millican: God does NOT exist.” Youtube Video, Posted by OxfordUnion December 21, 2012, 6:33

[vi] ““Does God exist?” David Wood vs. Michael Shermer (Christian vs. Atheist Debate)” Youtube Video, Posted by Acts17Apologetics, October 30, 2016 1:07:00